March 2015 - Wine-devel - list.winehq.org

Re: [2/3] windowscodecs: Implement PngDecoder_Block_GetCount.
by Alexandre Julliard 27 Mar '15

27 Mar '15

Vincent Povirk <madewokherd(a)gmail.com> writes: > From 398cd35190fddb05de97f6b125897be6c0e9e92d Mon Sep 17 00:00:00 2001 > From: Vincent Povirk <vincent(a)codeweavers.com> > Date: Wed, 25 Mar 2015 14:26:28 -0500 > Subject: [PATCH 2/3] windowscodecs: Implement PngDecoder_Block_GetCount. > > --- > dlls/windowscodecs/pngformat.c | 80 +++++++++++++++++++++++++++++++++++++++--- > 1 file changed, 76 insertions(+), 4 deletions(-) ../../../tools/runtest -q -P wine -T ../../.. -M windowscodecs.dll -p windowscodecs_test.exe.so metadata && touch metadata.ok metadata.c:949: Test succeeded inside todo block: GetCount failed, hr=80070057 metadata.c:952: Test succeeded inside todo block: GetCount failed, hr=0 metadata.c:953: Test succeeded inside todo block: unexpected count 1 Makefile:375: recipe for target 'metadata.ok' failed -- Alexandre Julliard julliard(a)winehq.org

1 0

Re: [PATCH 3/4] advapi32: Add DACL inheritance support in SetSecurityInfo
by Erich E. Hoover 26 Mar '15

26 Mar '15

I do not believe this is the correct place to make this change. Inheritance still needs to occur even when SetSecurityInfo is not used, which pretty much means it has to be within the wineserver itself so that the same code can be used for CreateFile. The code I put together for this (as written it is somewhat dependent on my other ACL code) can be found here: https://github.com/wine-compholio/wine-staging/tree/master/patches/server-I… The first patch is for CreateFile and the second one is for SetSecurityInfo. I'm happy to try and change this to help you solve your problem without my other ACL code, it's been on my TODO list anyway - I've just been too busy with work lately :/ Best, Erich

4 4

Re: urlmon: Remove unneeded NONAMELESSXXX directives.
by Jacek Caban 26 Mar '15

26 Mar '15

Hi Francois, On 03/26/15 15:29, Francois Gouget wrote: > Also put them in the files that need them. It may be a bit late for such question after so many patches already went in, but what's the point of this patch? Those defines are defined in central place, so that we don't have to deal with them in each .c file. Even if they are not needed for some .c files, they are harmless. BTW, I think we could even consider putting this macro in a central place like windef.h inside __WINESRC__ guard. Jacek

1 0

named memory-mapped file
by Leon Bogaert 26 Mar '15

26 Mar '15

Hi all, I don't know if this is the correct list. If it's not I'm sorry. I have kind of a weird question. I'm playing this game iRacing which exposed it's api via named memory map ( https://github.com/meltingice/node-iracing/blob/master/lib/irsdk_utils.cpp) Is there a way I can access that named memory-mapped file via Linux? I did some Googling and I don't think it's possible but I thought I made sure to ask around here. If that isn't possible I think I'll write a simple program that would run under wine and expose that data in a Linux-friendly manner. Leon

3 4

how to upload and view reports
by Shubham kumar 26 Mar '15

26 Mar '15

Sir, i have applied for winetest graphical user interface project. Problem is to see the test results in application itself we first need to know that how it shows report on test.winehq.org so that, we could get idea on how it shows report after uploading. But, on the website i could not find any link to upload the report. Where/How am i supposed to upload the report and view the test results on website?

2 1

Application for GSoc 2015 - Implement vbscript.dll builtins for Wine
by Shuai Meng 26 Mar '15

26 Mar '15

Hello everyone, I'm Shuai Meng, a student from Beihang University of China, who have attended Gosc 2014 for wine vbscript. I would like to continue to work on this module this year: on one hand, I am familiar with it through last year's experience; on the other hand, as I have studied it deeper and deeper, I find it more difficult than imagined, and I think understanding and fulfilling this interpreter of vbscript is very meaningful and interesting. Last year, the relatively simple built-in functions have been implemented by my effort, with the generous help of many warm-hearted people. The left functions are more challenging, I have to study more about COM and DLL in order to understand the mechanism. Now let's talk about the details of my plan, the functions unimplemented can be sorted as below: 1.The first kind is the functions about date and time, they have not been implemented because Wine's vbscript is not able to parse date constant in the form of "#yy-mm-dd#" yet. The functions below can be hopefully implemented as long as I sovle this problem: IsDate CDate Day Month Weekday Year Hour Minute Second DateValue TimeValue DateSerial TimeSerial DateAdd DateDiff DatePart. 2.The second kind is about strings, they are LenB LeftB RightB MidB InStrB Replace String StrComp. I have sent some patches to implement String, which all failed. Then I realized that I have to solve the problem of locale before I can implement them. 3.The third kind is about characters: Asc AscB ChrB ChrW Escape Unescape, I have already sent patch for Asc, but due to the same problem as strings, it didn't work. Again I have to implement GetLocale and SetLocale function, however as Piotr pointed, these two functions are not registered in native typelib, so I have to figure out another way to implement them, instead of using idl files. 4.The fourth kind is about arrays: IsArray LBound UBound Array Filter Join Split. Though haven't considered them seriously, but I can feel that they harder than those listed before. 5.The fifth kind is about math, but only left Randomize and Rnd, they are a little difficult, but I think it's unnecessary to implement them manually, I mean there must be a wrapper function. I have found some samples in jscript and other modules, I will learn from them. 6.The sixth kind is others: Eval Execute ExecuteGlobal GetRef LoadPicture InputBox FormatNumber FormatCurrency FormatPercent FormatDateTime, they are the most difficult, I am not sure whether I can implement them, but I will try. 7.The remaining functions are about Error handling: Err_Description Err_HelpContext Err_HelpFile Err_Source Err_Raise. I haven't studied them yet, but sooner or later I will meet them. This is my idea for GSoC 2015, I appreciate for any comment. Thank you.

1 0

Re: ntdll/tests: Add tests for GetFinalPathNameByHandleA/W.
by Marvin 26 Mar '15

26 Mar '15

Hi, While running your changed tests on Windows, I think I found new failures. Being a bot and all I'm not very good at pattern recognition, so I might be wrong, but could you please double-check? Full results can be found at https://testbot.winehq.org/JobDetails.pl?Key=12425 Your paranoid android. === wvistau64 (32 bit file) === file.c:369: Test failed: attributes of file "testfile.xxx" are 0x0820 === wvistau64 (64 bit file) === file.c:369: Test failed: attributes of file "testfile.xxx" are 0x0820

1 0

Re: winecfg: add note about 32 bit prefixes to dialog
by Stefan Dösinger 26 Mar '15

26 Mar '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I think making this visible on some way is a good idea. I have wondered about this myself, and so have many other users. > Context behind this: To run Simcity 2000 I change my Wine prefix > settings to mimic NT 3.5 so that it uses a different file chooser > dialog which doesn't crash the game (for whatever reason). This > last time installing things I was confused about why I didn't see > the option for NT 3.5 until slackner reminded me on IRC that I need > a 32 bit prefix to have that option. Somewhat offtopic to your patch: Is this a known bug in the game? Does it crash in Windows XP or newer unless you set Win9x or Win NT compatibility? It may be a good idea to investigate the actual crash. > So, I thought it might be useful if the dialog could have a note > explaining why those options aren't there if you're running from a > 64 bit prefix -- and thus this patch. To make room for the text I > made the applications Listbox 20 pixels shorter. - From a user interface point of view, what do you think about adding those old Windows versions in the listbox, maybe with a different text like "Windows 98 not available in 64 bit", and programming the code to set the dropdown field back to the last setting if the user decides to choose one of those settings? Or instead of having the modified text, pop up an error box and set the setting back. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBAgAGBQJVEea8AAoJEN0/YqbEcdMwQCsP/0Uw+ogXWRBtTZz46g8s6hAD wZjtYdFWfx9FMbITjpA9E2VLvmHj4tN6wZDanqOLuaHIRL3Mxa8eKaeZZ+LqE6pL mZWWJcWEQ+V8C/Fr6UTFUsbFhhSDTPhw9BJH/vZBof7uw0evNfBvV6lFQ+nf5bq7 Tpcwu+cBCeVlm9atvChZLwFAuBqjeVuH5ibyGnADIxKScHKvM9rbuJf2f3g3d7bW jCpR5sxoXfgeDeZYyaFh4o3WPHjVKriT3Twz2pYgHe/yZbxgoto72a7WNr5dMdHu 4/2FCQoBoDX2HahnCgWiL7fSDLk3pZFBQ78JRZUKDcS5nj868PszszAn4REVPXEW sgL5dgQqEJs0WhlhAePnUhuPudEaOIdKRDHqMftqJZpTwMxBwo16vnEnsLejrhlB Br2s2YIlqhpF0GKx4TeaefYFlJI/uRlxyhsmUBOvcfZ/BrLtpY9JMSHg0nGPnbI4 IUyU6iBQ5KSLjyDFDFTQ/gS19vbVtzJuEI88wtcUb0I2FknIa2zZN7GBDKN+VxOl hhtiiz1wgfxrrejVTxRKVVDF6bk5BpBWHLIyVIWVxZLiKy46I5Zie2MhbG74A5ks P1vph231JUHeea51wxFOfbUjw7LcTwhHswj8YM2I1EPizOH0AWJS5WQg0K+19o+g ch59pOw13mCF3Fl3LCtz =8FQi -----END PGP SIGNATURE-----

4 4

Re: [PATCH] advapi32: succeed for null pointer in ChangeServiceConfig2
by Nikolay Sivov 25 Mar '15

25 Mar '15

On 19.03.2015 15:06, Bernhard Übelacker wrote: > Hello, > attached patch avoids an issue when trying to run Flatout 2 with copy > protection. > > When starting Flatout2.exe for the first time it wants to install some > services, but shows in wine some error messages. > > The only visible sign in terminal output, that something went wrong, is > this: > err:rpc:I_RpcReceive we got fault packet with status 0x3e6 > > And of course later some dialog showing error 998 (=0x3e6) and a > unhandled exception. > > This is because the SC_RPC_CONFIG_INFOW parameter has in its > descr->lpDescription a null pointer instead of a valid LPWSTR. > > Right now this null pointer would be accessed in > svcctl_ChangeServiceConfig2W and causes an exception there , but is > hidden by the RPC layer. > > (And of course this makes the SF4 protection still not working.) > > Kind regards, > Bernhard > Is it actually called with infolevel 1?

3 9

Re: msvcrt: Fix _copysign bugs for NAN/INF(try 2)
by Henri Verbeet 25 Mar '15

25 Mar '15

On 25 March 2015 at 12:51, Zheng Chen <chanchengcc(a)gmail.com> wrote: > The last patch was damaged by my email client. > I would recommend everyone that doesn't already use it to spend a couple of minutes to learn git send-email. It avoids a lot of issues, and in general just saves everyone (not just reviewers) a lot of time.

2 1